4.2.2 inductive statistics 1 upa package 4, module 2 inductive statistics
TRANSCRIPT
14.2.2 Inductive Statistics
UPA Package 4, Module 2
INDUCTIVE STATISTICS
24.2.2 Inductive Statistics
Inductive Statistics
• Statistics and the Geographic Research Process• Normal Distribution• Inductive Statistics• Correlations and Relationships• Introduction exercise 4.2.2
34.2.2 Inductive Statistics
Statistics and the Geographic Research Process
44.2.2 Inductive Statistics
Frequency Distributions
0 5 10
Household members total
0
50
100
150
200
250
300
Fre
qu
ency
Mean = 5.16Std. Dev. = 2.319N = 1,510
Histogram
0.00 20000.00 40000.00 60000.00 80000.00 100000.00120000.00
Total Wage Income 2002
0
100
200
300
400
500
Fre
qu
ency
Mean = 6951.0865Std. Dev. = 9423.36206N = 855
54.2.2 Inductive Statistics
Normal Distribution
• Normal curve• Bell-shape curves described by its mean (center of the
distribution) and standard deviation (shape of the curve)
64.2.2 Inductive Statistics
Normal Distribution
• Areas under the normal curve• The 68 – 95 - 99.7% rule
• Z or standard scores
• Standard score =
74.2.2 Inductive Statistics
Normal Distribution
• Symmetrical distributions • Positive and negatively
skewed distributions • Mean not equal to median• Impact of extreme cases
e.g. very high incomes on mean
84.2.2 Inductive Statistics
Inductive Statistics
• Hypotheses
Null and Research Hypothesis
Assumptions• Level of significance
Two and one tailed test
Degree of Freedom• Tests
Parametric and non-parametric tests• Cross tables
94.2.2 Inductive Statistics
Inductive Statistics
Income
20000180001600014000120001000080006000
Hou
se v
alue
200
180
160
140
120
100
80
60
40
20
scattergram (house value / income)
104.2.2 Inductive Statistics
Inductive Statistics
• Tests• Level of significance• Cross tablesEA Slum Type * Slum/Non-Slum Cross tabulation
Slum/Non-Slum
Slum
households Non-slum
households Total
Count 548260 82700 630960 Slum % of Total 26.8% 4.0% 30.9% Count 252340 1161180 1413520
EA Slum Type
Non-slum % of Total 12.3% 56.8% 69.1% Count 800600 1243880 2044480 Total % of Total 39.2% 60.8% 100.0%
114.2.2 Inductive Statistics
Inductive Statistics
Procedure for testing statistical hypotheses:• Assumptions• Sampling procedure• Sign. Level, critical region• Testing• Decision/conclusions
124.2.2 Inductive Statistics
Correlations and Relationships
Correlations
House Value Income House Value 1.000 .815 Pearson Correlation
Income .815 1.000
HOUSES_$ . .000 Sig. (1-tailed)
TTLRES_$ .000 .
HOUSES_$ 87 87 N
TTLRES_$ 87 87
Correlation Coefficient 0.815
134.2.2 Inductive Statistics
Introduction Exercise 4.2.2
• Create Histograms• visual check on normality• Calculate z-scores and ‘areas under the normal curve’• Create a scattergram• Calculate correlation coefficient
144.2.2 Inductive Statistics
Introduction Exercise 4.2.2
Histogram / Bar chart
Earth/Sand/Mud
Mud m
ixed with dung
Coarse wood planks
Palm/bam
boo
Cement (not polished)
Parquet or polished wood
Linoleum
Ceramic tiles
Cement (polished)
Carpet
Terrazzo
0
100
200
300
400
500
600
Fre
qu
ency
What is the most common material of the floor in the living room?
0 10000 20000 30000
Income
0
500
1,000
1,500
Fre
qu
ency
Mean = 10027.12Std. Dev. = 1974.064N = 8,976
Median=9756.00
154.2.2 Inductive Statistics
Introduction to Exercise 4.2.2
Population density vs. Income
Data refer to mean values per municipality in the Netherlands
Higher Income – higher population densities?
8000 10000 12000 14000
income
0
2000
4000
6000
8000
10000
12000
po
p_d
n_1
R Sq Linear = 0.087